循环提取Tratamentos水平下响应变量均值SD并绘图的问题排查

需要循环提取Tratamentos各水平（"T-0"、"T-2"、"T-3"、"T-5"）下所有响应变量（"GE"、"PC"、"IVE"、"CE"、"AM"、"EE"、"FB"、"PB"）的均值和标准差（SD）并绘制图表，但循环代码报错：

Error: unexpected '=' in:
"  df_err<-ds.1.2%>%group_by(Tratamentos)%>%
    summarize(vars[j] ="

手动处理单个变量（如GE）可正常运行，求循环代码的修复方案。

原代码核心问题：

• dplyr的summarize函数无法直接用字符串变量作为新列名，语法不支持vars[j] = ...这种写法
• 调用mean()/sd()时直接传入字符串vars[j]，而非引用数据框对应的列，导致逻辑错误

# 加载依赖包
library(dplyr)
library(ggplot2)
library(ggpubr)

# 导入并预处理数据
ds.1.1 <- read.csv("https://raw.githubusercontent.com/Leprechault/trash/main/exp_ivan_1.csv")
ds.1.2 <- ds.1.1 %>% dplyr::select(!EM)

# 提取响应变量名
vars <- colnames(ds.1.2)[-1]
# 初始化图表存储列表
plots_pull <- vector("list", length(vars))

# 循环处理每个变量
for (j in seq_along(vars)) {
  current_var <- vars[j]
  
  # 计算各处理组均值
  df_err <- ds.1.2 %>%
    group_by(Tratamentos) %>%
    summarize(!!current_var := mean(.data[[current_var]], na.rm = TRUE)) %>%
    as.data.frame()
  
  # 计算各处理组标准误（SEM）
  df_sd_err <- ds.1.2 %>%
    group_by(Tratamentos) %>%
    summarize(sd = stats::sd(.data[[current_var]], na.rm = TRUE) / sqrt(sum(!is.na(.data[[current_var]])))) %>%
    as.data.frame()
  
  # 合并均值与标准误数据
  df_combined <- merge(df_err, df_sd_err, by = "Tratamentos")
  
  # 绘制带误差线的折线图
  pd <- position_dodge(0.1)
  plots_pull[[j]] <- ggplot(df_combined, aes(x = Tratamentos, y = .data[[current_var]])) +
    geom_errorbar(aes(ymin = .data[[current_var]] - sd, ymax = .data[[current_var]] + sd),
                  position = pd, width = 0.2) +
    geom_line(aes(linetype = Tratamentos), position = pd) +
    scale_colour_grey(start = 0, end = .5) +
    labs(y = current_var, title = current_var) +
    theme_bw() +
    theme(
      panel.border = element_blank(), 
      panel.grid.major = element_blank(),
      panel.grid.minor = element_blank(), 
      axis.line = element_line(colour = "black"),
      axis.text.x = element_text(angle = 0, hjust = 0),
      axis.text.y = element_text(angle = 0, hjust = 0),
      text = element_text(size = 18, family = "serif"),
      legend.position = "none"
    )
}

# 组合所有图表并输出
ggarrange(plotlist = plots_pull,
          font.label = list(size = 14, color = "black", face = "plain", family = "serif"),
          ncol = 2, nrow = 4,
          align = "hv",
          common.legend = TRUE, 
          legend = "bottom")

• 字符串变量引用：用!!current_var :=将字符串转为列名，.data[[current_var]]引用数据框对应列，解决dplyr的非标准求值问题
• 标准误计算：用sum(!is.na(.data[[current_var]]))获取每组有效样本量，避免NA值干扰样本数统计
• 列表初始化：提前创建与变量数量匹配的空列表，防止索引越界
• 数据合并：将均值和标准误数据合并，简化ggplot的数据调用逻辑
• 图表可读性：添加变量名作为y轴标签和标题，方便识别每个图表对应变量

内容的提问来源于stack exchange，提问作者Leprechault